estonian-mcp

Use this skill whenever the user is writing, editing, proofreading, or studying Estonian text and the estonian-mcp connector is available — even if they don't explicitly ask. Covers proofreading (spell_check first, then verify lemma and case forms via lemmatize and analyze_morphology on uncertain words), tone-aware rewrites (classify_register before and after edits to confirm the register matches the audience), vocabulary diversification (synonyms from WordNet for same-meaning swaps; find_related_words via fastText for adjacent concepts in marketing copy), morphology study, compound splits, syllabification, and named-entity extraction from Estonian news or documents. Core principle — never invent Estonian lemmas, case endings, conjugations, or spellings; always verify with the MCP tools, because models routinely hallucinate Estonian inflections. Documents tool quirks inline (fastText antonym-near-neighbours, polysemy, Vabamorf gaps on neologisms, synonyms-vs-related-words decision rule) and anti-patterns.